[PATCH V2 3/3] Adding device_dma_parameters->offset_preserve_mask to NVMe driver.

[Keith Busch]

On Mon, Feb 01, 2021 at 10:30:17AM -0800, Jianxiong Gao wrote:
> @@ -868,12 +871,24 @@ static blk_status_t nvme_map_data(struct nvme_dev *dev, struct request *req,
>  	if (!iod->nents)
>  		goto out_free_sg;
>  
> +	offset_ret = dma_set_min_align_mask(dev->dev, NVME_CTRL_PAGE_SIZE - 1);
> +	if (offset_ret) {
> +		dev_warn(dev->dev, "dma_set_min_align_mask failed to set offset\n");
> +		goto out_free_sg;
> +	}
> +
>  	if (is_pci_p2pdma_page(sg_page(iod->sg)))
>  		nr_mapped = pci_p2pdma_map_sg_attrs(dev->dev, iod->sg,
>  				iod->nents, rq_dma_dir(req), DMA_ATTR_NO_WARN);
>  	else
>  		nr_mapped = dma_map_sg_attrs(dev->dev, iod->sg, iod->nents,
>  					     rq_dma_dir(req), DMA_ATTR_NO_WARN);
> +
> +	offset_ret = dma_set_min_align_mask(dev->dev, 0);
> +	if (offset_ret) {
> +		dev_warn(dev->dev, "dma_set_min_align_mask failed to reset offset\n");
> +		goto out_free_sg;
> +	}
>  	if (!nr_mapped)
>  		goto out_free_sg;

Why is this setting being done and undone on each IO? Wouldn't it be
more efficient to set it once during device initialization?

And more importantly, this isn't thread safe: one CPU may be setting the
device's dma alignment mask to 0 while another CPU is expecting it to be
NVME_CTRL_PAGE_SIZE - 1.